No strategy wins every time. What separates those who survive from those who blow their account is not being right more often, but losing little when wrong and still standing when a bad streak arrives. These are the risk management rules, with verified numbers and the uncomfortable cases people tend to skip.
Risk a small, fixed percentage
Many traders risk between 0.5% and 2% of their balance per trade. It is not a magic rule, but it has a mathematical reason: after n losses in a row your account stands at (1 − risk)^n. After ten straight losses, 0.5% risk leaves you 4.9% down; 1%, 9.6%; 2%, 18.3%; 5%, 40.1%.
Losses are also asymmetric: losing 10% requires a 11.1% gain to recover; 20% requires 25%; 30% requires 42.9%; and 50% requires 100%. Small risk is not timidity, it is staying in the zone where recovery is still realistic.
Think in R multiples
Call R what you risk on a trade. Losing your stop is −1R; winning twice what you risked is +2R. Measuring in R lets you compare trades of different sizes and understand your performance independent of balance.
Example: six trades with results of +2R, −1R, −1R, +1.5R, −1R and +3R add up to +3.5R. If your R was 50 USD, you made 175 USD, and if you later move to an R of 100 USD, the same sequence would be worth 350 USD. Journal analysis in R is comparable over time; analysis in money is not.
Losing streaks: how many to expect, and how likely
Long losing streaks are not exceptional bad luck: they are statistics. The probability that your next n trades all lose is q^n, where q is your loss rate. With a 40% win rate (q = 0.6), five straight losses have a 7.8% chance at any given moment, and eight straight, 1.7%.
What matters is the chance of it happening at least once over a long period. With that 40% win rate, across 100 trades the probability of meeting at least one run of 8 straight losses is about 49%, and for 10 straight, about 20%. With a 50% win rate, a run of 6 losses shows up at least once in 100 trades with a 55% probability. If your 2% risk could not survive 10 straight losses (an 18.3% loss), assume it will happen.
Daily and weekly limits
Decide in advance how much you will lose in a day and in a week, and stop when you reach it. One possible scheme, at 1% risk per trade: a maximum of 3R lost per day and 6R per week. The worst day then costs about 3% and the worst week about 6%. The exact figures are yours; what matters is that they are written down before you open the chart.
The worst decisions are usually made right after a losing streak, trying to win it back with bigger size or more trades. A daily limit is not a punishment: it is a firewall against the emotional state in which analysis stops being objective. Stopping after two or three consecutive losses and reviewing what happened is a cheap and effective rule.
Correlation: how many trades you really have
Opening EUR/USD and GBP/USD in the same direction is not two independent trades: they often move together. Buying EUR/USD, GBP/USD and AUD/USD and selling USD/CHF is four positions but one idea: dollar weakness. At 1% each, your effective risk approaches 4% on a single bet.
Correlation is not constant: it tends to rise exactly in moments of stress, when it hurts most. Stock indices (S&P 500, Nasdaq) also move together. A practical rule is to treat correlated positions as one and split a single trade’s risk among them.
Account size shapes the plan
With a small account, 1% is only a few units of money and you run into the minimum lot problem: with 500 USD, 1% is 5 USD, and a wide stop can force you to risk double. The temptation is to raise the risk percentage so it “feels worth it”. That is the shortest road to ruin: asking a small account for the results of a large one can only be done by increasing risk.
With a large account, the issue shifts: positions are bulkier, slippage weighs more and concentration in a few instruments carries more risk. In both cases, the percentage per trade must be the variable you command, not the money you want to make this month.